Technical Surgical Failures: Presentation, Etiology, and Evaluation Carrie A

8 Technical Surgical Failures: Presentation, Etiology, and Evaluation Carrie A. Sims and David W. Rattner

Approximately 48,000 patients undergo anti- surgery, many perceive that residual symptoms reflux procedures each year in the United States. represent an indication of fundoplication Although surgery is the most effective treatment failure.It is well known,however,that symptoms for gastroesophageal reflux disease (GERD),anti- correlate poorly with the presence of acid reflux reflux operations have reported failure rates after fundoplication. Soper and Dunnegan1 between 3–30%.This wide variability reflects dif- found that 26% of those undergoing laparo- ferences in operative technique,differences in the scopic anti-reflux surgery reported postopera- length of reported follow-up, and differences in tive foregut symptoms. After an extensive the definitions used to describe failure. For the evaluation, 35% had no demonstrable abnor- purposes of this chapter, failure is defined as the mality and their symptoms resolved without development of recurrent or new symptoms after intervention.1 Galvani et al.2 studied 124 anti-reflux surgery combined with documented patients with persistent or recurrent foregut pathologic gastroesophageal reflux or anatomic symptoms after laparoscopic fundoplication. failure. Failures occurring within the first 3 Only 39% were found to have acid reflux by 24- months of surgery are termed early failures and hour pH monitoring. Viewed another way, two- are generally caused by technical errors.Diaphrag- thirds of the patients who were taking matic stressors such as coughing,straining,vom- acid-reducing medications postoperatively were iting,retching,and weight lifting increase the risk found to have normal 24-hour pH probes of recurrence, especially in the early postopera- studies (the studies were performed off med- tive period. When failures occur after 3 months, ication).2 Almost every patient experiences they are termed late failures and a combination some degree of dysphagia in the early postop- of factors may be responsible. The size of the erative period. In a review by Perdikis et al.,3 original hiatal hernia, increased intraabdominal dysphagia occurred in 20% of the 2453 patients pressure,the presence of Barrett’s esophagus,and analyzed. Initial dysphagia may be secondary to the use of steroids predispose to late failures.This distal esophageal edema or transient esophageal chapter will discuss the evaluation and man- dysmotility and most patients can be treated agement of failed fundoplications. expectantly. Given the disparity between symptoms and demonstrable anatomic or physio- Presenting Symptoms of logic abnormalities, documenting functional status with appropriate testing must be per- Failed Anti-Reflux Operations formed before ascribing symptoms after fundoplication to a failed operation. Patients with GERD often have associated Patients with failed anti-reflux surgery typi- gastrointestinal motility disorders. Because cally complain of dysphagia, heartburn, vomit- patients have high expectations of anti-reflux ing, or a combination of these symptoms.4 The

91 92

MANAGING FAILED ANTI-REFLUX THERAPY

Figure 8.1. Upright abdominal radiograph demonstrating a dilated gas-filled stomach consistent with the gas bloat syndrome. majority of symptomatic recurrences occur Patients with an improperly constructed fundo- within 2 years.5 Patients whose dysphagia per- plication may not be able to easily belch and sists for more than 3 months postoperatively painful abdominal bloating may arise when should be suspected of having an anatomic swallowed air is “trapped.” This is readily diag- problem. In the early postoperative period, sub- nosed with a plain film of the abdomen showing sternal chest pain or discomfort is another a distended gas-filled stomach (Figure 8.1) or in common symptom.Although the etiology is not the absence of an X-ray, prompt relief of pain by well understood, the pain may be secondary passage of a nasogastric tube. The “gas bloat to esophageal spasm, irritation from the syndrome” should be differentiated from the esophageal dissection and mobilization, or more common complaint of generalized referred pain from the crural repair. The pain abdominal bloating and increased flatulence as may be described as a dull ache although some the latter tends to resolve on its own over time. patients describe it as heartburn. Usually this, too, can be managed conservatively.Vomiting in the postoperative setting is very abnormal and Methods of Evaluation often signifies disruption of the fundoplication. More ominously, it may be the presenting sign Given the poor correlation between symptoms of an incarcerated iatrogenic paraesophageal and anatomic failure, a careful and thorough hernia.If a patient experiences severe chest pain evaluation is warranted. A complete history and in the setting of retching or straining, the diag- physical should be performed with particular nosis of a transhiatal herniation of the wrap attention to the patient’s current symptoms.Are should be considered. This is a surgical emer- the symptoms similar to those experienced gency and a water-soluble contrast study should before the original surgery? Do symptoms of be done immediately to confirm the diagnosis. reflux or dysphagia predominate? Was there a If herniation is present, the patient should be precipitating event? Do antacid medications returned expeditiously to the operating room ameliorate the symptoms? The patient’s original for a laparoscopic or open reduction of the operative report should be obtained to clarify herniated stomach. the type of fundoplication and extent of dissec- Whereas dysphagia and heartburn are the tion. Any prior preoperative radiographs and most common symptoms after fundoplication, physiologic test results should also be obtained rarely, patients may complain of “gas bloat” and reviewed. If the patient’s symptoms are characterized by the onset of severe epigastric identical to their prior symptoms of reflux, a 2- pain approximately 30 minutes after eating. week trial of omeprazole at 40mg/d should be 93

TECHNICAL SURGICAL FAILURES: PRESENTATION, ETIOLOGY,AND EVALUATION initiated. Symptoms that completely resolve on these changes may impact the decision to reop- this regimen should raise suspicion for recur- erate or treat medically. rent reflux. The patient can be offered a contin- Esophageal manometry should be routinely ued course of medical therapy as a reasonable performed before considering reoperation. option. Many patients feel so well after success- Manometry provides an objective means of ful anti-reflux surgery, however, that they prefer assessing the location and resting pressure of another operation to a lifetime of medical the lower esophageal sphincter. It can also therapy. If the patient does not respond to provide an assessment of the functional status omeprazole or has symptoms of dysphagia, the of esophageal peristalsis and sphincter relax- work-up should proceed with more invasive ation. Manometric studies are critical when monitoring and diagnostic studies in an attempt to elucidate the etiology of their symptoms. A barium swallow should be the initial diagnostic study in the work-up of any symptomatic patient. This relatively noninvasive, inexpensive study will define the patient’s anatomy and help clarify the relationship of the gastroesophageal junction to the hiatus. This study may also demonstrate gastroesophageal reflux and can detect evidence of delayed esophageal emptying. A barium swallow is particularly helpful when the patient presents with symptoms of dysphagia or pain and can help delineate a gross anatomic defect that might explain the patient’s symptoms (Figure 8.2). However, the failure to visualize reflux on a barium study does not exclude the possibility that the patient is expe- riencing pathologic reflux. Because patients may have symptoms consistent with reflux without evidence of gastroesophageal reflux, a 24-hour pH study is important in patients whose anatomy seems to be intact. This functional study confirms the presence of pathologic gastroesophageal reflux. By maintaining a 24- hour diary, the patient’s subjective assessment of reflux can be correlated with monitored episodes of reflux. Patients who have “reflux” symptoms, but a normal 24-hour pH study, are likely to have another cause for their symptoms and will not benefit from refundoplication. Upper gastrointestinal endoscopy should be routinely performed in evaluating patients who are symptomatic after a fundoplication. Endoscopy and barium swallows provide complementary information. In up to 10% of patients, an endoscopy will reveal an anatomic problem not appreciated by a barium swallow.6 In particular, endoscopic evaluation may reveal a “spiraling” or “twisting” of the wrap that may be missed by standard barium studies (Figure 8.3). Endoscopy also helps assess complications of gastroesophageal reflux such as esophagitis Figure 8.2. A barium swallow demonstrating a slipped Nissen and Barrett’s mucosal changes. The degree of fundoplication. 94

MANAGING FAILED ANTI-REFLUX THERAPY

Failure can occur at the esophageal, wrap, or crural level, although there may be overlapping or concurrent issues. Before the wide adoption of laparoscopic techniques,wrap disruption was the most common mode of failure. In the laparoscopic era, the most common cause of failure is herniation of the wrap through the diaphragmatic hiatus. The construction of a fundoplication (particularly a 360° fundoplication) may unmask previously unrecognized esophageal dysmotility or misdiagnosed achalasia leading to severe postoperative dysphagia. Chronic inflammation can also contribute to esophageal failure. Both Barrett’s esophagus and severe esophageal reflux are associated with chronic esophageal Figure 8.3. A retroflexed endoscopic view of the gastroe- inflammation. Chronic inflammation results in sophageal junction demonstrating a twisted fundoplication. fibrosis, foreshortening, esophageal dysmotility, and poor acid clearance. Poor acid clearance in turn contributes to more esophageal irrita- evaluating the patient who presents with dys- tion and the vicious cycle is propagated. Over phagia, as these patients may have a previously time, the esophagus may become significantly undiagnosed esophageal motility disorder. It foreshortened and fibrotic. Although there may be particularly difficult to differentiate is controversy over the true incidence of the patients with misdiagnosed achalasia from short esophagus, we believe that this entity those whose fundoplication is too tight causing exists. secondary poor esophageal peristaltic function. A variety of issues involving fundoplication Moreover, patients who initially had normal construction can contribute to failed anti-reflux esophageal function before surgery may surgery (Figure 8.4). The easiest failure to develop secondary achalasia after fundoplica- diagnose and repair is the “missin’ Nissen”—a tion.7 If reoperative surgery is indicated, the fundoplication that is disrupted or completely type of fundoplication chosen may depend on undone. A “slipped” Nissen results when the the results of esophageal manometry. Patients body of the stomach intussuscepts through the complaining of dysphagia who are found to fundoplication. This creates an hourglass defect have poor esophageal motility probably should with part of the stomach residing above the not be offered a 360° wrap. wrap and part below. Patients with a “slipped” Patients with persistent bloating, nausea, fundoplication often experience severe reflux vomiting, abdominal pain, and early satiety and regurgitation because the pouch of stomach should undergo gastric emptying studies. These above the wrap traps food and serves as a reser- symptoms may be secondary to previously voir of acid-rich refluxate below an incompetent undiagnosed gastroparesis. An injury to the esophageal sphincter. Similarly, a wrap may be vagus nerves may also lead to abnormal gastric misplaced around the upper stomach rather function with rapid emptying of liquids and than around the esophagus. This creates an delayed emptying of solids. If gastroparesis is hourglass defect in which the wrap is below the detected, the success rate of a reoperation is diaphragmatic hiatus, but the upper stomach lower and a pyloroplasty should be performed. and gastroesophageal junction are above the diaphragm.Another common error particularly in the laparoscopic era is use of the body or even Potential Causes of Failure antrum of the stomach to construct a Nissen fundoplication (Figure 8.5). This leads to a Regardless of the surgical nuances, failed anti- twisted, bulky wrap that fails to function prop- reflux operations can be analyzed and subdi- erly. Lastly, a fundoplication that is too tight vided into three distinct anatomic regions. may result in dysphagia. Since the work of 95

TECHNICAL SURGICAL FAILURES: PRESENTATION, ETIOLOGY,AND EVALUATION

Dunnington and DeMeester8 established the laparoscopic fundoplication decreased from efficacy of the floppy fundoplication, most sur- 19% to 4%. Whereas others have demonstrated geons construct 360° wraps over a 56–60 French that division of the short gastric vessels does not dilator to avoid this problem. However, con- improve the clinical outcome of laparoscopic structing a wrap over a large dilator without fundoplication, the Nissen procedure per- adequate fundic mobilization can still lead to formed in this study as the control was not the tension. By routinely dividing the short gastric classic “floppy” fundoplication with full mobi- vessels and approximating the crura, Soper and lization.9 As such, we believe that the short Dunnegan1 reported the failure rate of primary gastric vessels should be divided with full mobi-

Figure 8.4. Types of surgical failure of Nissen fundoplication. (Reprinted from Hinder RA. Gastroesophageal reﬂux disease. In: Bell RH Jr, Rikkers LF,Mulholland MW, eds. Digestive Tract Surgery: A Text and Atlas. Philadelphia: Lippincott-Raven Publishers; 1996:19, with permission.) 96

MANAGING FAILED ANTI-REFLUX THERAPY

hiatal closure5 we find this method both inaccu- rate and dangerous. The quality of the crura and ability to obtain a well-approximated and tension-free closure are essential. Patients with large hiatal hernias at the time of their initial surgery are three times more likely to develop a recurrence.1 Recent publications have demonstrated the feasibility and utility of judiciously placing prosthetic material to buttress the crural closure when the crural fibers are attenuated.10,11 Without a doubt, the best time to prevent recurrence is at the time of the original procedure. Anticipating potential pitfalls and prob- lems at the esophageal, wrap and crural level Figure 8.5. The body or antrum of the stomach can be mistak- during the initial procedure will prevent later enly used to form the fundoplication.A proper fundoplication is complications. We recommend esophageal constructed by wrapping “A” around the distal esophagus and manometry for all patients before anti-reflux bringing “B” anterior to the esophagus to join “A.” By bringing surgery and if weak peristalsis is present, a “C”anteriorly to complete the wrap,a malformed fundoplication partial, rather than a total fundoplication, will be created. should be performed. The esophagus should be adequately mobilized such that 2–3cm of tension-free intraabdominal esophagus can be lization of the fundus in order to create a wrap obtained. If a foreshortened esophagus is dis- that lies comfortably around the esophagus. covered preoperatively or intraoperatively, an There are a myriad of partial fundoplications esophageal lengthening procedure should be in use today. The most common laparoscopic performed. The establishment of a 2- to 4-cm partial fundoplication is the posterior fundopli- length of intraabdominal esophagus is a funda- cation described by Andre Toupet. Because few mental principle of anti-reflux surgery. If surgeons had experience with this repair in the tension is required to keep a fundoplication in open era, modern-day laparoscopic surgeons the abdomen, transdiaphragmatic herniation tend to make this fundoplication too short. In will ultimately result. The crura should always contrast to the Nissen fundoplication, longer is be closed with a nonabsorbable suture, often better for a Toupet procedure. The wrap should reinforced with the use of pledgets. extend for at least 4cm. Belsey fundoplication failures are usually attributed to inadequate esophageal mobilization or improper depth of Treatment Options suture placement when constructing the wrap. The competency of the crural closure is crit- Appropriate treatment of failed anti-reflux ical in the performance of a successful fundo- surgery may range from reassurance to re- plication. The crural closure can either be too operative therapy. Revisional surgery can be tight leading to dysphagia or too lax leading to recommended when the preoperative evalua- transdiaphragmatic herniation of the wrap. In tion identifies a surgically correctable problem open operations, the crural closure should corresponding to the patient’s symptoms. In admit the tip of the surgeon’s index finger general, operations that have failed for technical snugly when a nasogastric tube lies in the reasons can be corrected by a second operation. esophagus. Obviously, this rule of thumb cannot The most appropriate surgical approach will be used for laparoscopic operations. We try to depend on the patient’s previous operation and leave 1–1.5cm of space between the anterior the results of the preoperative evaluation (see border of the esophagus and the anterior Chapter 11). margin of the hiatus to approximate the degree For patients with persistent dysphagia, of closure obtained in open operations. esophageal dilation should be the first line of Although some use a bougie to calibrate the therapy. Often, multiple dilations can loosen a 97

TECHNICAL SURGICAL FAILURES: PRESENTATION, ETIOLOGY,AND EVALUATION tight, but properly oriented wrap. However, if Patients whose initial anti-reflux surgery was dysphagia is caused by a tight crural closure, performed via laparotomy are more difficult to dilation usually will not work and reoperation approach laparoscopically than those whose is often necessary. Similarly, if the wrap is mal- primary procedure was laparoscopic. If the positioned, conservative therapy is unlikely to reoperation fails, patients may become eso- provide benefit. Patients whose dysphagia fails phageal cripples with irreparable motility dis- to respond to 2 or 3 dilations should be sus- orders and face the prospect of esophageal pected of having a poorly constructed wrap or replacement surgery. Hence, the stakes are an overly tight crural closure. Reoperation extremely high in the reoperative setting. should be considered if symptoms persist for In general, patients who have previously more than 3–4 months. had open surgery should have open revisional surgery to avoid unnecessary risk and pro- longed operative times. When reoperating on What Is the Best Surgical Approach patients with longstanding reflux disease, it is to a Reoperation? essential that the surgeon have the capability to perform an esophageal lengthening procedure. The most important factor a surgeon should Large iatrogenic paraesophageal hernias can be consider in choosing the surgical approach for very difficult to reduce laparoscopically. Finally, reoperation is the likelihood that he or she can in obese male patients, laparoscopic exposure of perform a safe and technically proper recon- the scarred hiatus may be difficult and one struction of the anti-reflux mechanism. Irre- should consider a transthoracic approach (Table spective of the surgical approach and choice of 8.1). If a patient is deemed not to be a candidate operation, the surgeon must establish a 3- to for a laparoscopic reoperation, the choice of 4-cm segment of intraabdominal esophagus, transabdominal versus transthoracic approach improve the lower esophageal sphincter resting depends on the perceived need for esophageal pressure, and reestablish a valve mechanism at mobilization/lengthening and body habitus. A the gastroesophageal junction. Multiple studies thoracotomy provides the best opportunity to have demonstrated the safety and efficacy of fully mobilize the esophagus up to the level of laparoscopic reoperations, but all these re- the aortic arch and overcomes difficulty expos- ports come from high-volume experienced ing the hiatus in obese patients. However, if a centers.5,12–16 Whereas some surgeons are transthoracic approach is chosen, the abdomen extremely skilled in laparoscopic techniques, must be prepped and draped into the operative others are less facile. Laparoscopic reopera- field. Difficult reoperations often require a com- tions are clearly more difficult than primary bined thoracic and abdominal approach to anti-reflux procedures and the potential for dissect the hiatus and mobilize the fundus if it serious complications (e.g., unrecognized per- is heavily scarred from prior surgery. Although foration of the stomach and esophagus, vagal the peritoneal cavity can be accessed through a nerve injury) should not be underestimated. counter incision in the diaphragm, extending a

Table 8.1. Choosing a laparoscopic or open approach for redo fundoplication. Factor Laparoscopic Approach Laparotomy Thoracotomy Prior repair via laparotomy ØÆÆ Prior laparoscopic repair ≠ÆÆ Large hiatal hernia Æ* Ø≠* Obesity ÆØ≠ Concern of short esophagus Æ* Ø≠*

≠=good choice. Æ=no contraindication. Ø=poor choice. * Must have skills to perform lengthening procedure. 98

MANAGING FAILED ANTI-REFLUX THERAPY thoracotomy incision across the costal margin crural fibers toward the anterior margin of the often provides the best exposure to deal with hiatus, the hiatal orifice is effectively displaced particularly difficult cases. cephalad. This transposition of the hiatal orifice lengthens the intraabdominal segment of Choosing a Partial esophagus because the anterior portion of the hiatus is cephalad to the posterior portion of the or a Total Fundoplication hiatus. If the gastroesophageal junction lies at The surgical literature is replete with articles the level of the hiatal closure,one must do some- debating the pros and cons of the Nissen thing to achieve an adequate intraabdominal fundoplication. Advocates of partial fundo- segment of esophagus. The first step should be plications such as the Toupet and Belsey esophageal mobilization. This can be done tran- fundoplication point out advantages of less dys- shiatally or transthoracically. If the segmental phagia and preservation of the ability to vomit. arteries to the esophagus are divided to the level Proponents of the Nissen fundoplication claim of the aortic arch and the vagal branches to superior control of acid reflux as well as ease of the hilum of the lungs are divided, one can gen- performance of the procedure. In fact, there is erally gain 2cm of esophageal length. If the little level 1 evidence to support the superiority intraabdominal esophageal segment is still of one procedure over another when perform- inadequate, there are several methods for ing a redo fundoplication. The choice of fundo- lengthening the intraabdominal segment (see plication should be tailored to the symptom or Chapter 14). anatomic defect needing correction. Patients The Collis gastroplasty is the most widely who had a good short-term result from a Nissen used technique to lengthen the esophagus. First fundoplication should probably have a full wrap described in conjunction with transthoracic reconstructed. Those patients who had a partial hiatal hernia repairs, the Collis gastroplasty can fundoplication with poor control of acid reflux also be performed with minimally invasive should be considered for conversion to a Nissen. approaches. Although some have reported out- If a clear technical error can be identified that standing long-term results with the Collis 17 the neoesophagus may contain caused a full fundoplication to fail, one should gastroplasty, acid-secreting mucosa causing concern that not hesitate to reconstruct the 360° fundoplica- patients with Barrett’s esophagus may continue tion in a proper manner. However, it is logical to to be exposed to acidic irritation. In the current perform partial fundoplications on patients that era wherein most reoperations follow failed have had a prior Nissen fundoplication and laparoscopic Nissen fundoplication, the proxi- complain of persistent dysphagia or gas bloat mal end of the Collis gastroplasty may become syndrome. Patients who undergo reoperation to ischemic because the short gastric vessels have correct wrap herniation may benefit from a pro- been previously divided. This may result in a cedure that anchors the fundus to the hiatus stricture that is very difficult to treat by dilation. such as a Hill repair,Belsey procedure,or Toupet In 1996, Swanstrom et al.18 described a mini- procedure. When an esophageal lengthening mally invasive transthoracic Collis gastroplasty procedure is performed, a partial fundoplica- technique (see Chapter 13). For this approach, a tion has theoretical advantages because the 12-mm trocar is placed in the right anterior gastric tube that becomes the distal neoesopha- axillary line in the third or fourth intercostal gus is aperistaltic. space.A 35-mm tissue stapler is introduced into the right chest and passed along the posterior Considerations for Esophageal medial sulcus until it can be seen laparoscopi- Lengthening Procedures cally from the abdomen indenting the medi- astinal pleura. The pleura is incised and the It is essential that the gastroesophageal junction stapler is advanced parallel to the esophagus. lie tension free in the abdomen before creating A 46- to 48-French bougie is advanced into a fundic wrap. The length of tension-free the stomach along the lesser curvature. While intraabdominal esophagus should be measured the fundus is retracted laterally, the stapler is after closing the crural defect. When the crura advanced along the bougie, adjacent to the angle are closed from the caudal condensation of the of His. The stapler is fired creating a 3-cm 99

TECHNICAL SURGICAL FAILURES: PRESENTATION, ETIOLOGY,AND EVALUATION gastric tube. The crura are closed in standard manner and a fundoplication is performed (Figure 8.6). A Collis gastroplasty can also be performed laparoscopically without violating the thoracic cavity (see Chapter 12).An esophageal bougie is advanced along the lesser curvature. A circular stapling device is used to create a “buttonhole” in the gastric fundus adjacent to the bougie. A 35-mm tissue stapler is passed into the “buttonhole” and advanced parallel to the bougie toward the angle of His. The linear stapler is then fired creating a neoesophagus19 (Figure 8.7). The introduction of roticulating endoscopic staplers has greatly simplified laparoscopic esophageal lengthening procedures. Many surgeons now resect a wedge-shaped segment of the fundus to create a neoesophagus rather than using the buttonhole technique described above. Once the short gastric vessels have been divided and with a bougie in the esophagus, a linear stapler is fired across the

Figure 8.7. This laparoscopic method of addressing the foreshortened esophagus requires two types of staplers.The anvil of a circular stapler is brought through and through the body of the stomach, following a stitch on a straight needle (A). The stapler is ﬁred, creating an aperture through the stomach (B). A linear cutting stapler is ﬁred from this aperture to the gastroesophageal junction, completing the lengthening gastroplasty. (Reprinted from Horvath et al.,22 with permission.)

fundus near the gastroesophageal junction, per- pendicular to the esophagus. This then permits the surgeon to divide the fundus adjacent to the dilator and parallel to the esophagus creating the neoesophagus. A fundoplication is then performed. Management of the Difficult Hiatus Closing the crura in reoperative surgery can be challenging. The first and most important step in repairing hiatal defects is to avoid destroying the crural fibers while performing the initial dissection of the area. Spending the extra time Figure 8.6. A minimally invasive transthoracic method of creating a Collis gastroplasty entails visualization of the hiatus from to dissect this area carefully will be rewarded above. After the proximal stomach is mobilized, a linear stapler later in the operation when it becomes time to is brought through the thoracic port and fired alongside a close the hiatus. On rare occasions, the crura bougie in the stomach (B). This creates a lengthening gastro- will be very fibrotic creating esophageal plasty (C) that is used to form a fundoplication (D). (Reprinted obstruction at the hiatal level. This problem is from Horvath et al.,22 with permission.) easily remedied by dividing a portion of the 100

MANAGING FAILED ANTI-REFLUX THERAPY crus. The most common problem, however, is Technical Tips for Laparoscopic dealing with a large hiatal defect. In most instances, primary closure can be accomplished. Reoperations The stoutest crural fibers are the posterior fibers A five trocar technique described for primary so sutures should be placed deeply to encom- laparoscopic fundoplication can be used for pass them. In laparoscopic reoperations, the redo surgery (see Chapter 12).Adhesiolysis may intraabdominal pressure should be lowered to 8 be challenging and is best accomplished with or 10mmHg in order to diminish the diaphrag- cold scissors to avoid thermal or conductive matic stretch. injury to the esophagus and vagi. The liver is There are situations in which primary closure almost invariably stuck to the site of the previ- seems impossible.When this occurs, the options ous fundoplication and needs to be freed as the are to make a relaxing incision in the diaphragm first step of the procedure. As with primary and close this defect with prosthetic material or surgery, the crura are identified and dissected. to place prosthetic material directly into the In general, the left crus is more easily identified. hiatus. Surgical dogma has been that placement If the short gastric vessels have not been taken of prosthetic material in the hiatus would lead previously, they should be divided to facilitate to erosion of the foreign body into the esopha- identification of the left crus. The right crus can gus. In the laparoscopic era, however, there are be more challenging and is best isolated by numerous reports claiming both the safety and finding the caudate lobe and proceeding superi- benefit of using prosthetic material for difficult orly and to the left. The crura are dissected in a hiatal closures. Although there is relatively 360° manner such that a Penrose drain can be limited follow-up, the use of mesh at the passed behind the esophagus (or stomach if esophageal hiatus has been associated with there is a herniated fundoplication). Once the significantly reduced recurrence rates with hiatus has been delineated clearly, attention is minimal morbidity. Laparoscopic refundoplica- turned to the prior fundoplication.We routinely tion with a circular polypropylene mesh was take down the previous fundoplication, try to performed in 24 patients with intrathoracic her- restore normal anatomy, and then reconstruct niation of the wrap. Although one patient had the fundoplication again. The retained sutures severe dysphagia requiring pneumatic dilation are divided sharply and the fundus is peeled postoperatively, all patients had good to excel- away from the stomach and esophagus circum- lent functional outcome at 1 year follow-up.11 ferentially. Sharp dissection is continued in Prosthetic material may be useful in reinforcing order to avoid inadvertent injury to the vagi. the crural closure, particularly if the hiatal dis- Special attention is required along the left lateral ruption is large or if the tissue is less than wall of the esophagus to ensure that the prior robust. The use of polytetrafluoroethylene mesh fundoplication is fully mobilized. in conjunction with a Nissen fundoplication was With the fundoplication taken down, an investigated in patients with a hiatal defect assessment of the intraabdominal esophageal >8cm. With at least 1 year follow-up, this length is made. When necessary additional prospective, randomized controlled study esophageal length is obtained as previously demonstrated that 8 of the 36 patients undergo- described. The crura are reapproximated with ing simple cruroplasty developed recurrences, nonabsorbable pledgetted sutures starting whereas none of the 36 patients with polyte- caudally and progressing cephalad toward the trafluoroethylene mesh recurred.20 This esophageal hiatus. Generous crural bites should benefit of prosthetic reinforcement has been be taken. If closing the hiatus is difficult, tension observed in multiple other studies.10,20,21 on the diaphragm can be reduced by lowering Although there has been concern regarding the the intraabdominal insufflation pressure. If the possibility of erosion of the mesh into the crural closure remains under tension, mesh stomach or esophagus, these fears have not reinforcement should be considered. Attention materialized—at least in the short term. Most is turned to the recreation of the fundoplication. recently, biodegradable small intestinal submu- A 56-French esophageal bougie is passed orally cosal patches have become available and appear into the stomach. The fundus should easily pass to hold promise as an adjunct to closing large posterior to the esophagus and its orientation hiatal defects. should be confirmed by the “shoe shine”test.For 101

TECHNICAL SURGICAL FAILURES: PRESENTATION, ETIOLOGY,AND EVALUATION a Nissen fundoplication, the wrap should not be of esophageal functions tests. Arch Surg 2003;138: under any tension. Nonabsorbable pledgetted 514–519. 3. Perdikis G, Hinder RA, Lund RJ, Raiser F, Katada N. sutures are used to approximate the fundus Laparoscopic Nissen fundoplication: where do we around the distal 2–3cm of esophagus with each stand? Surg Laparosc Endosc 1997;7:17–21. suture incorporating a generous purchase of 4. Carlson MA, Frantzides CT. Complications and results stomach and esophagus. By anchoring fundo- of primary minimally invasive anti-reflux procedures: a plication to the esophagus, migration is less review of 10,735 reported cases. J Am Coll Surg 2001; 193:428–439. likely. Further steps to prevent herniation of the 5. Hinder RA, Klinger PJ, Perdikis G, Smith SL. Manage- fundoplication include fixation of the posterior ment of the failed anti-reflux operation. Surg Clin North portion of the wrap to the crural closure and Am 1997;77:1083–1098. performance of an anterior gastropexy or 6. Hunter JG. Approach and management of patients with recurrent gastroesophageal reflux disease. J Gastrointest gastrostomy. Surg 2001;5:451–457. 7. Stylopoulos, Bunker CJ, Rattner DW. Development of achalasia secondary to laparoscopic Nissen fundoplication. J Gastrointest Surg 2002;6:368–376. Conclusion 8. Dunnington GL, DeMeester TR. Outcome effect of adherence to operative principles of Nissen fundoplica- Patients who experience technical failures after tion by multiple surgeons. The Department of Veterans anti-reflux surgery typically complain of dys- Affairs Gastroesophageal Reflux Disease Study Group. phagia, heartburn, vomiting, or a combination Am J Surg 1993;166:654–657. 9. O’Boyle CJ, Watson DI, Jamieson GG, Myers JC, Game of these symptoms. Understanding why the PA, Devitt PG. Division of short gastric vessels at laparo- original procedure failed and establishing the scopic Nissen fundoplication: a prospective double- physiologic basis of the patient’s symptoms are blind randomized trial with 5 year follow up. Ann Surg critical elements in choosing management 2002;235:165–170. 10. Carlson MA, Richards CG, Frantzides CT. Laparoscopic options. Evaluation typically includes a careful prosthetic reinforcement of hiatal herniorrhaphy. Dig history and a complete physical examination, Surg 1999;16:407–410. barium swallow, upper endoscopy, esophageal 11. Ganderath FA, Kamolz T, Schweiger UM, Pointer R. manometry, esophageal pH monitoring, and Laparoscopic refundoplication with prosthetic hiatal often an assessment of gastric emptying. closure for recurrent hiatal hernia after primary failed anti-reflux surgery. Arch Surg 2003;138:902–907. Symptom recurrence in the absence of impor- 12. Bais JE, Horbach JMLM, Masclee AAM, Smout AJPM, tant anatomic abnormalities can often be Terpstra J,Gooszen HG.Surgical treatment for recurrent managed medically. When a technical failure gastro-oesophageal reflux disease after failed anti-reflux has occurred that results in symptoms that are surgery. Br J Surg 2000;87:243–249. 13. Floch NR, Hinder RA, Kingler PJ, et al. Is laparoscopic difficult to control or an important anatomic reoperation for failed anti-reflux surgery feasible? Arch abnormality, reoperation is necessary. Choosing Surg 1999;134:733–737. the right approach to reoperation requires 14. Granderath FA, Kamolz T, Schweiger UM. Long-term honest appraisal of the surgeon’s experience and follow-up after laparoscopic refundoplication for failed capabilities as well as tailoring the operative anti-reflux surgery: quality of life, symptomatic outcome and patient satisfaction. J Gastrointest Surg procedure to solve the patient’s physiologic 2002;6: 812–818. abnormality. Because the stakes are high in 15. Pointer R, Bammer T, Then P, Kamolz T. Laparoscopic reoperative esophageal surgery, consideration refundoplications after failed anti-reflux surgery. Am J should be given to referring such cases to Surg 1999;178:541–544. 16. Watson DI, Jamieson GG, Game PA, Williams RS, Devitt surgeons or centers doing a high volume of anti- PG. Laparoscopic reoperation following failed anti- reflux surgery. reflux surgery. Br J Surg 1999;86:98–101. 17. Luketich JD, Grondin SC, Pearson FG. Minimally invasive approaches to acquired shortening of the esophagus: laparoscopic Collis-Nissen gastroplasty. Semin References Thorac Cardiovasc Surg 2000;12:173–178. 18. Swanstrom LL, Marcus DR, Galloway GQ. Laparoscopic 1. Soper NJ, Dunnegan D. Anatomic fundoplication failure Collis gastroplasty is the treatment of choice for the after laparoscopic anti-reflux surgery. Ann Surg 1999; shortened esophagus. Am J Surg 1996;171:477–481. 229:669–677. 19. Johnson AB, Oddsdottir M, Hunter JG. Laparoscopic 2. Galvani C, Fisichella PM, Gorodner MV,Perretta S, Patti Collis gastroplasty and Nissen fundoplication: a new MG. Symptoms are a poor indicator of reflux status after technique for the management of esophageal foreshort- fundoplication for gastroesophageal reflux disease: role ening. Surg Endosc 1998;12:1055–1060. 102

MANAGING FAILED ANTI-REFLUX THERAPY

20. Frantzides CT, Madan AK, Carlson MA, Stavropoulos 22. Horvath KD, Swanstrom LL, Jobe BA. The short esoph- GP. A prospective, randomized trial of laparoscopic agus: pathophysiology, incidence, presentation, and polytetrafluoroethylene (PTFE) patch repair vs simple treatment in the era of laparoscopic anti-reflux surgery. cruroplasty for large hiatal hernia. Arch Surg 2002;137: Ann Surg 2000;232:630–640. 649–653. 23. Hunter JG, Smith CD, Branum GD, et al. Laparoscopic 21. Basso N, De Leo A, Genco A, et al. 360 degrees laparo- fundoplication failures: patterns of failure and response scopic fundoplication with tension-free hiatoplasty in to fundoplication revision. Ann Surg 1999;230:595–604. the treatment of symptomatic gastroesophageal reflux disease. Surg Endosc 2000;14:164–169.